🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: demon.tech.modems
Article #18122

Re: Ascend connection problems

#18122
From: Jim Segrave
Date: Thu, 21 Nov 1996 00:00
60 lines
4025 bytes
In article <NW3KqGAro5kyEAx+@loncps.demon.co.uk>,
Chris Salter  <chris@loncps.demon.co.uk> wrote:
> Surely what we should be interested in is the actual output of the 
> ascend units is, not what the manufacturer says it should be? Have you 
> measured the output across a sample of individual units? A sample of 
> USR's should also be measured.

We can't measure at the ouput, as it's directly into the E1 from the
Telco - it's only analogue for a very short time.
 
> Also if the output level is under software control, could you confirm 
> that all your ascend units are at the same software level?

They were until the 33.6 lines went into beta-test. They are using a
later software revision.

> I seem to remember that during the implementation of the 100 per cent 
> coverage back in Nov/Dec of last year that one (or a batch?) Ascend unit 
> had to be withdrawn because of connection problems and that it was at a 
> later software level than the others. My own current problem seems 
> identical to that I experienced at that time. My memory tells me that it 
> was associated with output levels but I could be wrong. I unfortunately 
> did not keep Demons newsgroup article explaining what the problem was.

We've had to remove a couple of interim releases when testing showed
problems, but that simply meant reloading a previous release. There
have been several upgrades, normally done on a test Ascend with
dialups from Demon, followed by testing on one or two Ascends in one
of the hunt groups. The only times I can recall having to drop back to
previous releases weren't output evel problems as far as I can recall,
but I'm not part of the team which actually does the installation and
testing of new Ascend releases.

> Are the Ascend units assembled in the UK or are they shipped direct from 
> the US? dB levels were never my strong point but I think US modems 
> output at 0 dB whereas the BABT would require them to output at -14 dB. 
> However, reading the USR Courier manual I note that it has max output at 
> -13 dB for private wire and -9 dB for PSTN. I assume this may indicate 
> that my own recollection is associated with private wires (leased 
> lines). I would interpret the statement that "Later releases of software 
> provide a way of adjusting the transmit level, but only downwards" as 
> indicating that the Ascends are probably shipped pre-configured for 
> maximum output levels within any particular PTT's spec. All this does 
> raise the possibility of units being shipped out of spec if Ascend QC 
> isn't quite on the ball. I'm not suggesting that this is the case, just 
> that it is a possibility.

They are assembled in the US. The output level is given as -13 (dbm
presumably) and can only be adjusted downwards. I do not know offhand
what the BABT or US requirements are, although Ascend has stated that
the default setting of -13 is the maximum they are allowed to use in
the UK.

> Finally, just for my own curiosity, how many incoming (logical) 
> connections does a single Ascend unit support?

60 connections/Ascend. The previous versions had 6 cards with 8 modems
each, the newer ones have 12 modems/card. Two E1s give a maximum of 30
connections/E1 (2 channels are used up for signalling).
Technical Details
Message-ID: <199611211713.RAA21699@jes.noc.nl.demon.net>
Path: arc1-ca.pugleaf.net!unknown!not-for-mail
References: <32828191.4B53@helion.demon.co.uk> <VUVrVCALjRjyEwbp@tuppambr.demon.co.uk> <199611181957.TAA07529@jes.noc.nl.demon.net> <NW3KqGAro5kyEAx+@loncps.demon.co.uk>

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 102a94,4dfa1137aae35630
X-Google-Attributes: gid102a94,public
X-Google-Thread: fad9d,4dfa1137aae35630
X-Google-Attributes: gidfad9d,public
From: Jim Segrave <jes@demon.net>
Subject: Re: Ascend connection problems
Date: 1996/11/21
Message-ID: <199611211713.RAA21699@jes.noc.nl.demon.net>
X-Deja-AN: 197847092
references: <32828191.4B53@helion.demon.co.uk> <VUVrVCALjRjyEwbp@tuppambr.demon.co.uk> <199611181957.TAA07529@jes.noc.nl.demon.net> <NW3KqGAro5kyEAx+@loncps.demon.co.uk>
x-mail2news-user: jes@demon.net
x-mail2news-path: jes.noc.nl.demon.net
organization: Demon Internet
reply-to: jes@demon.net
newsgroups: demon.tech.modems,demon.service